Step 5:
Load the layers you just created using the deepNetworkDesigner
clear;
clc;
close all;
mydata = imageDatastore('MerchData', ... % load data from folder named 'MerchData'
'IncludeSubfolders',true,
...
% Also include the subfolders (there are 5
of these for the 5 objects, cap, cube, ...)
'LabelSource','foldernames');
% The name of the subfolders supply the "correct answer" labels
[mydataTrain,mydataValidation] = splitEachLabel(mydata,0.7);
%Resize images in the image datastores to match the pretrained network GoogLeNet.
mydataResizedTrain = augmentedImageDatastore([224 224],mydataTrain,'colorpreprocessing','gray2rgb');
mydataResizedValidation = augmentedImageDatastore([224 224],mydataValidation,'colorpreprocessing','gray2rgb');
% colorpreprocessing option above makes sure that if your data set includes
% both color and grayscale images there won't be a problem loading them to the datastore.
% Color images usually have a dimension of mxnx3 where as grayscales are mxnx1
load MerchNewLayers %network created for transfer learning using deepNetworkDesigner & GoogleNet